Jun Bai
ABSTRACT
Digital breast tomosynthesis, or 3D mammography, has advanced the field of breast imaging diagnosis. It has been rapidly replacing the traditional full-field digital mammography because of its diagnostic superiority. However, automatic detection of breast cancer using digital breast tomosynthesis images has remained challenging, mainly due to their high resolution, high volume, and complexity. In this study, we developed a novel model for more precise detection of cancerous 3D mammogram images. The proposed model first, represents 3D mammograms as graphs, then employs a self-attention graph convolutional neural network model to effectively and efficiently learn the features of 3D mammograms, and finally, using the extracted features, identifies the cancerous 3D mammograms. We trained and evaluated the performance of the proposed model using public and private datasets. We compared the performance of the proposed model with those of multiple state-of-the-art CNN-based models as baseline models. The results show that the proposed model outperforms all the baseline models in terms of accuracy, precision, sensitivity, F1, and AUC.
- Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurelien Lucchi, Pascal Fua, and Sabine Süsstrunk. 2010. Slic superpixels. Technical Report.Google Scholar
- David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, and Lars Petersson. 2021. Graph-based deep learning for medical diagnosis and analysis: past, present and future. Sensors 21, 14 (2021), 4758.Google ScholarCross Ref
- Jun Bai, Russell Posner, Tianyu Wang, Clifford Yang, and Sheida Nabavi. 2021. Applying deep learning in digital breast tomosynthesis for automatic breast cancer detection: A review. Medical image analysis 71 (2021), 102049.Google ScholarCross Ref
- Mateusz Buda, Ashirbani Saha, Ruth Walsh, Sujata Ghate, Nianyi Li, Albert Swiecicki, Joseph Y. Lo, Jichen Yang, and Maciej Mazurowski. 2020. Breast Cancer Screening - Digital Breast Tomosynthesis (BCS-DBT). Type: dataset. Google ScholarCross Ref
- Cătălina Cangea, Petar Veličković, Nikola Jovanović, Thomas Kipf, and Pietro Liò. 2018. Towards sparse hierarchical graph classifiers. arXiv preprint arXiv:1811.01287 (2018).Google Scholar
- Stefano Ciatto, Nehmat Houssami, Daniela Bernardi, Francesca Caumo, Marco Pellegrini, Silvia Brunelli, Paola Tuttobene, Paola Bricolo, Carmine Fantò, Marvi Valentini, Stefania Montemezzi, and Petra Macaskill. 2013. Integration of 3D digital mammography with tomosynthesis for population breast-cancer screening (STORM): a prospective comparison study. The Lancet Oncology 14, 7 (June 2013), 583--589. Google ScholarCross Ref
- Cathy Coleman. 2017-05-01. Early Detection and Screening for Breast Cancer. Seminars in Oncology Nursing 33, 2 (2017-05-01), 141--155. Google ScholarCross Ref
- Michaël Defferrard, Xavier Bresson, and Pierre Vandergheynst. 2016. Convolutional neural networks on graphs with fast localized spectral filtering. Advances in neural information processing systems 29 (2016).Google Scholar
- Sergei V Fotin, Yin Yin, Hrishikesh Haldankar, Jeffrey W Hoffmeister, and Senthil Periaswamy. 2016. Detection of soft tissue densities from digital breast tomosynthesis: comparison of conventional and deep learning approaches. In Medical Imaging 2016: Computer-Aided Diagnosis, Vol. 9785. SPIE, San Diego, California, United States, 228--233.Google Scholar
- Hongyang Gao and Shuiwang Ji. 2019. Graph u-nets. In international conference on machine learning. PMLR, 2083--2092.Google Scholar
- Julianne S. Greenberg, Marcia C. Javitt, Jason Katzen, Sara Michael, and Agnes E. Holland. 2014. Clinical Performance Metrics of 3D Digital Breast Tomosynthesis Compared With 2D Digital Mammography for Breast Cancer Screening in Community Practice. American Journal of Roentgenology 203, 3 (June 2014), 687--693. Publisher: American Roentgen Ray Society. Google ScholarCross Ref
- Brian M Haas, Vivek Kalra, Jaime Geisel, Madhavi Raghu, Melissa Durand, and Liane E Philpotts. 2013. Comparison of tomosynthesis plus digital mammography and digital mammography alone for breast cancer screening. Radiology 269, 3 (2013), 694--700.Google ScholarCross Ref
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770--778.Google ScholarCross Ref
- Dae Hoe Kim, Seong Tae Kim, and Yong Man Ro. 2016. Latent feature representation with 3-D multi-view deep convolutional neural network for bilateral analysis in digital breast tomosynthesis. In 2016 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 927--931.Google ScholarDigital Library
- Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907 (2016).Google Scholar
- Boris Knyazev, Graham W Taylor, and Mohamed Amer. 2019. Understanding attention and generalization in graph neural networks. Advances in neural information processing systems 32 (2019).Google Scholar
- Junhyun Lee, Inyeop Lee, and Jaewoo Kang. 2019. Self-attention graph pooling. In International conference on machine learning. PMLR, 3734--3743.Google Scholar
- Xin Li, Genggeng Qin, Qiang He, Lei Sun, Hui Zeng, Zilong He, Weiguo Chen, Xin Zhen, and Linghong Zhou. 2020. Digital breast tomosynthesis versus digital mammography: integration of image modalities enhances deep learning-based breast mass classification. Eur Radiol 30, 2 (Feb. 2020), 778--788. Google ScholarCross Ref
- Thomas P Matthews, Sadanand Singh, Brent Mombourquette, Jason Su, Meet P Shah, Stefano Pedemonte, Aaron Long, David Maffit, Jenny Gurney, Rodrigo Morales Hoil, et al. 2020. A multisite study of a breast density deep learning model for full-field digital mammography and synthetic mammography. Radiology: Artificial Intelligence 3, 1 (2020), e200015.Google Scholar
- Elizabeth S. McDonald, Andrew Oustimov, Susan P. Weinstein, Marie B. Synnestvedt, Mitchell Schnall, and Emily F. Conant. 2016. Effectiveness of Digital Breast Tomosynthesis Compared With Digital Mammography: Outcomes Analysis From 3 Years of Breast Cancer Screening. JAMA Oncol 2, 6 (June 2016), 737--743. Publisher: American Medical Association. Google ScholarCross Ref
- Kayla Mendel, Hui Li, Deepa Sheth, and Maryellen Giger. 2019. Transfer Learning From Convolutional Neural Networks for Computer-Aided Diagnosis: A Comparison of Digital Breast Tomosynthesis and Full-Field Digital Mammography. Academic Radiology 26, 6 (June 2019), 735--743. Google ScholarCross Ref
- Ian G Murphy, Mary F Dillon, Ann O' Doherty, Enda W McDermott, Gabrielle Kelly, Niall O'higgins, and Arnold DK Hill. 2007. Analysis of patients with false negative mammography and symptomatic breast carcinoma. Journal of surgical oncology 96, 6 (2007), 457--463.Google ScholarCross Ref
- Kyung Jin Nam, Boo-Kyung Han, Eun Sook Ko, Ji Soo Choi, Eun Young Ko, Dong Wook Jeong, and Ki Seok Choo. 2015. Comparison of full-field digital mammography and digital breast tomosynthesis in ultrasonography-detected breast cancers. The Breast 24, 5 (Oct. 2015), 649--655. Google ScholarCross Ref
- Alejandro Rodriguez-Ruiz, Jonas Teuwen, Suzan Vreemann, Ramona W. Bouwman, Ruben E. van Engen, Nico Karssemeijer, Ritse M. Mann, Albert Gubern-Merida, and Ioannis Sechopoulos. 2018. New reconstruction algorithm for digital breast tomosynthesis: better image quality for humans and computers. Acta Radiologica (Stockholm, Sweden : 1987) 59, 9 (Sept. 2018), 1051. Publisher: SAGE Publications. Google ScholarCross Ref
- Ravi K. Samala, Heang-Ping Chan, Lubomir M. Hadjiiski, Mark A. Helvie, Caleb Richter, and Kenny Cha. 2018. Evolutionary pruning of transfer learned deep convolutional neural network for breast cancer diagnosis in digital breast tomosynthesis. Phys. Med. Biol. 63, 9 (April 2018), 095005. Google ScholarCross Ref
- Richard E. Sharpe, Shambavi Venkataraman, Jordana Phillips, Vandana Dialani, Valerie J. Fein-Zachary, Seema Prakash, Priscilla J. Slanetz, and Tejas S. Mehta. 2015. Increased Cancer Detection Rate and Variations in the Recall Rate Resulting from Implementation of 3D Digital Breast Tomosynthesis into a Population-based Screening Program. Radiology 278, 3 (Oct. 2015), 698--706. Publisher: Radiological Society of North America. Google ScholarCross Ref
- Sadanand Singh, Thomas Paul Matthews, Meet Shah, Brent Mombourquette, Trevor Tsue, Aaron Long, Ranya Almohsen, Stefano Pedemonte, and Jason Su. 2020. Adaptation of a deep learning malignancy model from full-field digital mammography to digital breast tomosynthesis. In Medical Imaging 2020: Computer-Aided Diagnosis, Vol. 11314. International Society for Optics and Photonics, 1131406.Google Scholar
- Per Skaane, Andriy I. Bandos, Loren T. Niklason, Sofie Sebuødegård, Bjørn H. Østerås, Randi Gullien, David Gur, and Solveig Hofvind. 2019. Digital Mammography versus Digital Mammography Plus Tomosynthesis in Breast Cancer Screening: The Oslo Tomosynthesis Screening Trial. Radiology 291, 1 (Feb. 2019), 23--30. Publisher: Radiological Society of North America. Google ScholarCross Ref
- Per Skaane, Sofie Sebuødegård, Andriy I. Bandos, David Gur, Bjørn Helge Østerås, Randi Gullien, and Solveig Hofvind. 2018. Performance of breast cancer screening using digital breast tomosynthesis: results from the prospective population-based Oslo Tomosynthesis Screening Trial. Breast Cancer Res Treat 169, 3 (June 2018), 489--496. Google ScholarCross Ref
- Robert A Smith, Debbie Saslow, Kimberly Andrews Sawyer, Wylie Burke, Mary E Costanza, W Phil Evans III, Roger S Foster Jr, Edward Hendrick, Harmon J Eyre, and Steven Sener. 2003. American Cancer Society guidelines for breast cancer screening: update 2003. CA: a cancer journal for clinicians 53, 3 (2003), 141--169.Google Scholar
- American Cancer Society. 2020. How Common Is Breast Cancer? Breast Cancer Statistics. https://www.cancer.org/cancer/breast-cancer/about/how-common-is-breast-cancer.htmlGoogle Scholar
- Pizer Stephen M., Amburn E. Philip, Austin John D., Cromartie Robert, Geselowitz Ari, Greer Ari, ter Haar Romeny Bart, Zimmerman John B, and Karel Zuiderveld. 1987. Adaptive histogram equalization and its variations. Computer Vision, Graphics, and Image Processing 39 (Sept. 1987), 355--368. Google ScholarDigital Library
- László Tabár, Bedrich Vitak, Tony Hsiu-Hsi Chen, Amy Ming-Fang Yen, Anders Cohen, Tibor Tot, Sherry Yueh-Hsia Chiu, Sam Li-Sheng Chen, Jean Ching-Yuan Fann, Johan Rosell, Helena Fohlin, Robert A. Smith, and Stephen W. Duffy. 2011. Swedish Two-County Trial: Impact of Mammographic Screening on Breast Cancer Mortality during 3 Decades. Radiology 260, 3 (Sept. 2011), 658--663. Google ScholarCross Ref
- Du Tran, Heng Wang, Lorenzo Torresani, Jamie Ray, Yann LeCun, and Manohar Paluri. 2018. A closer look at spatiotemporal convolutions for action recognition. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. 6450--6459.Google ScholarCross Ref
- Srinivasan Vedantham, Andrew Karellas, Gopal R. Vijayaraghavan, and Daniel B. Kopans. 2015. Digital Breast Tomosynthesis: State of the Art. Radiology 277, 3 (Nov. 2015), 663--684. Publisher: Radiological Society of North America. Google ScholarCross Ref
- Zonghan Wu, Shirui Pan, Fengwen Chen, Guodong Long, Chengqi Zhang, and S Yu Philip. 2020. A comprehensive survey on graph neural networks. IEEE transactions on neural networks and learning systems 32, 1 (2020), 4--24.Google ScholarCross Ref
- Bin Yang, Haiwei Pan, Jieyao Yu, Kun Han, and Yanan Wang. 2019. Classification of medical images with synergic graph convolutional networks. In 2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW). IEEE, 253--258.Google ScholarCross Ref
- Mina Yousefi, Adam Krzyżak, and Ching Y. Suen. 2018. Mass detection in digital breast tomosynthesis data using convolutional neural networks and multiple instance learning. Computers in Biology and Medicine 96 (May 2018), 283--293. Google ScholarCross Ref
- Haochen Zhang, Dong Liu, and Zhiwei Xiong. 2019. Two-stream action recognition-oriented video super-resolution. In Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE, 8799--8808.Google ScholarCross Ref
- Yu Zhang, Xiaoqin Wang, Hunter Blanton, Gongbo Liang, Xin Xing, and Nathan Jacobs. 2019. 2d convolutional neural networks for 3d digital breast tomosynthesis classification. In 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, 1013--1017.Google ScholarCross Ref
- Yixiao Zhang, Xiaosong Wang, Ziyue Xu, Qihang Yu, Alan Yuille, and Daguang Xu. 2020. When Radiology Report Generation Meets Knowledge Graph. arXiv:2002.08277 [cs, eess] (Feb. 2020). http://arxiv.org/abs/2002.08277 arXiv: 2002.08277.Google Scholar
- Yu-Dong Zhang, Suresh Chandra Satapathy, David S. Guttery, Juan Manuel Górriz, and Shui-Hua Wang. 2021. Improved Breast Cancer Classification Through Combining Graph Convolutional Network and Convolutional Neural Network. Information Processing & Management 58, 2 (March 2021), 102439. Google ScholarCross Ref
Recommendations
CNN-Based CAD for Breast Cancer Classification in Digital Breast Tomosynthesis
ICGSP '18: Proceedings of the 2nd International Conference on Graphics and Signal ProcessingDigital breast tomosynthesis (DBT) is a promising new technique for breast cancer diagnosis. DBT has the potential to overcome the tissue superimposition problems that occur on traditional mammograms for tumor detection. However, DBT generates numerous ...
Digital Breast Tomosynthesis Parenchymal Texture Analysis for Breast Cancer Risk Estimation: A Preliminary Study
IWDM '08: Proceedings of the 9th international workshop on Digital MammographyStudies with mammograms have demonstrated a relationship between parenchymal texture and breast cancer risk. Although promising, texture analysis in mammograms is limited by the effect of tissue superimposition. Digital Breast Tomosynthesis (DBT) is a ...
Detection of Masses in Digital Breast Tomosynthesis Mammography: Effects of the Number of Projection Views and Dose
IWDM '08: Proceedings of the 9th international workshop on Digital MammographyWe are developing a computer-aided detection (CAD) system for masses on digital breast tomosynthesis mammograms (DBT). A data set of DBT from patients was collected with informed consent using a GE prototype DBT system. The system has an Rh/Rh x-ray ...
Comments